In the management of IT systems and other systems where large amounts of performance data is generated, there is a need to be able to gather, organize and store large amounts of performance data and rapidly search it to evaluate management issues. For example, server virtualization systems have many virtual servers running simultaneously. Management of these virtual servers is challenging since tools to gather, organize, store and analyze data about them are not well adapted to the task.
One prior art method for remote monitoring of servers, be they virtual servers or otherwise, is to establish a virtual private network between the remote machine and the server to be monitored. The remote machine to be used for monitoring can then connect to the monitored server and observe performance data. The advantage to this method is that no change to the monitored server hardware or software is necessary. The disadvantage of this method is the need for a reliable high bandwidth connection over which the virtual private network sends its data. If the monitored server runs software which generates rich graphics, the bandwidth requirements go up. This can be a problem and expensive especially where the monitored server is overseas in a data center in, for example, India or China, and the monitoring computer is in the U.S. or elsewhere far away from the server being monitored.
Another method of monitoring a remote server's performance is to put an agent program on it which gathers performance data and forward the gathered data to the remote monitoring server. This method also suffers from the need for a high bandwidth data link between the monitored and monitoring servers. This high bandwidth requirement means that the number of remote servers that can be supported and monitored is a smaller number. Scalability is also an issue.
Other non IT systems generate large amount of data that needs to be gathered, organized, stored and searched in order to evaluate various issues. For example, a bridge may have thousands of stress and strain sensors attached to it which are generating stress and strain readings constantly. Evaluation of these readings by engineers is important to managing safety issues and in designing new bridges or retrofitting existing bridges.
Once performance data has been gathered, if there is a huge volume of it, analyzing it for patterns is a problem. Prior art systems such as performance tools and event log tools use relational databases (tables to store data that is matched by common characteristics found in the dataset) to store the gathered data. These are data warehousing techniques. SQL queries are used to search the tables of time-series performance data in the relational database.
Several limitations result from using relational databases and SQL queries. First, there is a ripple that affects all the other rows of existing data as new indexes are computed. Another disadvantage is the amount of storage that is required to store performance metric data gathered by the minute regarding multiple attributes of one or more servers or other resources. Storing performance data in a relational database engenders an overhead cost not only in time but also money in both storing it and storing it in an indexed way so that it can be searched since large commercial databases can be required if the amount of data to be stored is large.
Furthermore, SQL queries are efficient when joining rows across tables using key columns from the tables. But SQL queries are not good when the need is to check for patterns in values of columns in a series of adjacent rows. This requires custom programming in the form of “stored procedures” which extract the desired information programmatically. This is burdensome, time consuming and expensive to have to write a custom program each time a search for a pattern is needed. As the pattern being searched for becomes more complex, the complexity of the stored procedure program also becomes more complex.
The other way of searching for a pattern requires joining the table with itself M-1 number of times and using a complex join clause. This becomes impractical as the number of joins exceeds 2 or 3.
As noted earlier, the problems compound as the amount of performance data becomes large. This can happen when, for example, receiving performance data every minute from a high number of sensors or from a large number of agents monitoring different performance characteristics of numerous monitored servers. The dataset can also become very large when, for example, there is a need to store several years of data. Large amounts of data require expensive, complex, powerful commercial databases such as Oracle.
There is at least one prior art method for doing analysis of performance metric data that does not use databases. It is popularized by the technology called Hadoop. In this prior art method, the data is stored in file systems and manipulated. The primary goal of Hadoop based algorithms is to partition the data set so that the data values can be processed independent of each other potentially on different machines thereby bring scalability to the approach. Hadoop technique references are ambiguous about the actual processes that are used to process the data.
Therefore, a need has arisen for an apparatus and method to reduce the amount of performance data that is gathered so that more sensors or servers can be remotely monitored with a data link of a given bandwidth. There is also a need to organize and store the data without using a relational database and to be able to search the data for patterns without having to write stored procedure programs, or do table joins and write complex join clauses.